Treatment Interruption & Affect on CD4 & Viral Load

Mike Saag presented a talk on an analysis he did by going back to his clinical database at UAB and looked for patients who stopped HAART in the last 4 years and then re-started it to see what the consequences were on CD4 and viral load. They looked at the patient records and charts for all patients seen in his clinic from Jan '96 (the start of the HAART era). Treatment interruption was defined as any length of time „ 30 days followed by HAART therapy for at least 30 days post interruption. They looked at 180 days of follow-up. Success was defined as the ability to get back to within 0.3 log of their baseline viral load or to get back to 90% of their CD4 count prior to treatment interruption.

There were 78 individuals in this analysis from a total population of 1246 persons; median age 38; 82% male; 70% caucasian. Saag said these were advanced patients who had a median of 3 regimens prior to the interruption, the highest viral load ever recorded on this group of interruption patients was a median of 137,000 copies/ml, and the lowest CD4 ever was 85 cells.

The median time on therapy prior to interruption was 506 days (range 101-1315), the treatment interruption itself was 67 days (range 31-1018), and the follow-up on average was 173 days (range 35-266). The median viral load prior to treatment interruption was 8,900 copies/ml (range 24-1 million), and the median CD4 was 230 (range 4-1203). The median viral load and CD4 after interruption was 404 (the best viral load reached) and 230. The time to reach the best viral load was 77 days (32-259), and the time to the best CD4 was 95 days (32-261).

Saag reported the last recorded median viral load and CD4 for this group during the follow-up period were 2100 copies/ml and 205 cells.  But he added there was no significant difference between the best values the patients attained and the different numbers during follow-up. He said "although there was a slight difference in numbers you could regain control to a fair degree and sustain it for the most part".

RESULTS:

CD4- 59% were able to regain 90% or better the CD4 count they had prior to the interruption.  Over 50% achieved better CD4s in the follow-up period. But 8 patients (10%) were unable to get back to 50% of their CD4 count prior to the interruption.

Viral load- 77% of the patients were able to get back to within 0.3 log or better of their viral load prior to the interruption. 72% were below their pre- interruption viral load. Meaning of course that 23% did not get back to within 0.3 log of viral load prior to interruption.

Using a univariate analysis to look at a number of predictors for success or failure to re-establish viral load and CD4, Saag was unable to find any of significance: length of the holiday, 3 of prior regimens, length of follow-up, duration of interruption/HAART (although there was a trend in favor of this as a predictor), pre-interruption CD4, pre-interruption viral load, end of interruption CD4, end of interruption viral load.

Saag reported the reasons for interruption in these patients: GI symptoms 36, metabolic complications 14 (elevated lipids, diabetes, lactic acidosis), virologic failure 11, financial (ran out of resources to pay for meds), neuropathy 6, CNS symptoms 5, non-compliance 5, other drug toxicity 5, new OI 1 (PCP). Some patients had more than 1 reason.

Saag reported on the 11 patients of the 78 that had virologic failure (median 31,000 copies/ml and CD4 of 125) prior to the treatment interruption. 8 of these 11 had a successful CD4 outcome and 10/11 had a successful viral load outcome. After the treatment interruption their median CD4 was 290 and viral load was 293. 13/78 resumed the same regimen they were using prior to the interruption but the others.

Saag concluded that most patients stopped therapy due to toxicity. Patients who had virologic failure had similar outcomes as those who stopped due to toxicity.